Microsoft Azure

Microsoft Azure (Azure) is Microsoft’s cloud computing platform that provides compute resources, data storage, databases, networking services, and other IT capabilities through a global infrastructure of data centers. Organizations can build and scale IT systems in the cloud without having to purchase and maintain all of the underlying physical hardware themselves.

Azure is one of the largest public cloud platforms, alongside AWS and Google Cloud. It is used for hosting virtual servers, applications, and databases, storing and backing up data, analytics, AI/ML, software development, and building hybrid IT infrastructure.

The platform includes services across different cloud models, from IaaS (Infrastructure as a Service) to managed platform and serverless solutions.

How Is Azure Infrastructure Organized?

Microsoft Azure’s global infrastructure is divided into regions — geographic areas where the platform’s data centers and cloud resources are located. When deploying many Azure services, customers select a region based on factors such as latency, cost, data residency requirements, and service availability.

Some regions provide Availability Zones — physically separate groups of data centers with independent power, cooling, and networking infrastructure. They allow workloads to be distributed so that an issue in one zone does not necessarily make the entire application unavailable.

Azure also provides mechanisms for replication and redundancy across regions to support geographically distributed systems. The specific high-availability and disaster recovery options depend on the cloud service being used.

Key Microsoft Azure Services

Azure includes a large number of services covering different layers of IT infrastructure. Some provide virtualized resources with extensive customer control, while others automate a significant portion of infrastructure administration.

Key Azure services include:

• Azure Virtual Machines — virtual Windows and Linux servers;
• Azure Blob Storage — object storage for large volumes of unstructured data;
• Azure Managed Disks — managed block storage for virtual machines;
• Azure SQL Database — a managed relational database service;
• Azure Virtual Network (VNet) — logically isolated virtual networks;
• Azure Functions — serverless code execution;
• Azure Kubernetes Service (AKS) — a managed Kubernetes environment;
• Microsoft Entra ID — a cloud-based identity and access management service.

For example, an enterprise application can run on Azure Virtual Machines, use Azure SQL Database as its database, and be separated from the rest of the network through an Azure Virtual Network.

Azure also provides tools and services for monitoring, cybersecurity, DevOps, analytics, IoT, backup, disaster recovery, and artificial intelligence.

IaaS, PaaS, and Serverless in Azure

Azure is more than a virtual server rental platform. It supports several cloud service models with different levels of responsibility for infrastructure management.

With IaaS, customers receive virtual compute, networking, and storage resources while retaining significant responsibility for managing the operating system and applications. Azure Virtual Machines are a typical example.

PaaS (Platform as a Service) allows customers to transfer more responsibility for the underlying infrastructure and software platform to Microsoft. For example, developers can deploy applications without having to manage the entire server environment themselves.

With a serverless model, such as Azure Functions, customers can run code without directly managing virtual servers. The underlying computing infrastructure still physically exists, but provisioning and scaling are handled by the cloud platform.

Where Is Azure Used?

Azure is particularly common among organizations that already use Microsoft enterprise technologies. Integration with Windows Server, Microsoft Entra ID, Microsoft 365, SQL Server, and other products makes it possible to connect on-premises and cloud infrastructure.

Common use cases include:

• hosting virtual servers and enterprise applications;
• storing, backing up, and processing data;
• running web applications and APIs;
• using managed databases;
• software development and testing;
• building Kubernetes infrastructure;
• analytics and AI/ML workloads;
• disaster recovery and secondary infrastructure.

For example, a company can move its test environments to Azure and run them only for the duration of a development cycle, while production systems continue to operate unchanged.

Azure is also widely used in hybrid cloud environments. An organization can keep some systems in its own data center while moving other workloads to Azure and connecting both environments through networking and management tools.

How Are Azure Resources Priced?

Many Azure services use a pay-as-you-go model, where costs depend on the resources actually consumed. However, the specific pricing mechanism varies by service.

Charges may be based on:

• compute resource usage or runtime;
• amount of data stored;
• storage and database operations;
• network data transfer;
• number of requests or executions;
• selected performance and redundancy tiers.

For example, Azure Functions is billed by the number of executions, while the cost of a virtual machine depends on its runtime and the configuration assigned to it.

For some resources, Azure also offers reservation and usage commitment options that can reduce costs compared with continuously paying standard consumption rates. As a result, total Azure costs depend on system architecture and workload characteristics rather than simply on the number of running virtual machines.

Azure and On-Premises Infrastructure

When a system is hosted in Azure, Microsoft is responsible for the physical infrastructure of the cloud platform. However, this does not mean that all responsibility for security and operations automatically shifts to the provider.

Like other major cloud platforms, Azure follows a Shared Responsibility Model. The division of responsibilities depends on the type of service. With IaaS, customers are responsible for more components, including the guest operating system and applications. With managed PaaS and SaaS services, Microsoft assumes responsibility for additional layers of the technology stack.

Organizations migrating to Azure must therefore continue to manage access, configurations, data, and other components that remain within their area of responsibility.

FAQ

🠔 Back to glossary